迪维西亚(Divisia)货币总量VAR模型的预测分析

The Forecasting Performance of the Divisia Monetary Aggregates VAR Models

本文针对最新编制的我国迪维西亚(Divisia)货币总量(指数)数据,运用向量自回归(VAR)模型预测实际产出和物价指数并与简单加总货币总量所得到的结果进行比较。结果显示,在预测我国实际产出值的VAR模型中,含有Divisia狭义货币(M1)总量的模型比含有简单加总货币总量M1的模型更精确;在预测我国的物价指数值的VAR模型中,含有Divisia广义货币(M2)总量的模型比含有简单加总货币总量M1和M2的模型更精确。

With the latest establishment of China＇s Divisia monetary aggregates data, we use four four - variable vector Autoregressive （VAR） models containing four kinds of monetary aggregates （M1, M2, D1 and D2） to predict the value of real output and price index, and then compare the results obtained from the model with Divisia and simple sum monetary aggregates. The results show that, in predicting the real output value of China, the VAR model with Divisia M1 monetary aggregates （D1） has more accurate result than the simple sum monetary aggregates M1 and M2, while in predicting the value of the price index, the VAR model with Divisia M2 monetary aggregates （D2） performs better than the simple sum monetary aggregates M1 and M2 counterparts